The sprayings get counted. So do the patients. Both numbers are going up

The City of Skopje has been counting treatments since 7 May, and the opposition has been counting patients. This week the West Nile death toll reached four, and what used to be fewer than one case a week became twelve in seven days. Meanwhile, the thirteenth round of spraying in Skopje went ahead while patients were already lying in the infectious diseases ward, with another twenty sprayings planned before the season ends.

The entire argument collapsed into a calendar - who sprayed when, how many times, under which report. What is missing is the only thing worth measuring: what actually changed on the ground after any of those sprayings. The inspectorate is now checking whether municipalities really sprayed the way the paperwork says they did, and testing jumped from forty a year to eight a day once the disease was already here. A count of treatments is not proof of effect. It is proof of activity.

When the check finally worked, it came from outside

The most convincing inspection of the week was not carried out by a single one of our institutions. A Bulgarian laboratory found formetanate above the legal limit in 11.6 tonnes of Macedonian peppers - the third European alert since 2023. The question nobody in charge wants to open is not who pays to destroy the shipment, but what happened to the rest of that harvest, the part that never went for export.

In the same week, Tabanovce revealed 33,722 fines that were never collected, because a foreign licence plate was a way out of the system. When that exit closed, the first arrest was made by a plate-reading camera, not by a patrol car. The mechanism works. It just was not ours - not the lab, and until a few days ago, not the habit of checking either.

Drawn up is not the same as built

The City of Skopje mapped the three spots that flood after every heavy rain and drew up technical solutions. Drew up - did not build. The distance between those two verbs is the whole story of this week.

The same pattern shows up everywhere. Bus line 37 came back on a new route, but with four departures and a six-hour gap in the middle of the day - a timetable for high-school students, not for a neighbourhood. The e-application for the university exists and is filled in online, after which the student walks it over in person in forty-degree heat. Every one of these has a finished document standing in for an unfinished job.

The fire still sets the timetable

This week the Army deployed preventively for the first time this summer - but only after thirty-seven fires were reported in twenty-four hours and the blaze near Čanište split into four fronts. Prevention that arrives in the third month of the season is still a reaction, just with a better name.

The clearest picture came from Kumanovo, which puts out its neighbours' fires without receiving a single denar from them. The crews go out, the work gets done, and the bill stays with whoever moved first. That is not a system - that is goodwill holding the place where a system should be.

The file exists. The name on it does not

After five years, three prosecutors and not one indictment, the modular hospitals case has been closed for lack of evidence. A closed case is not an acquittal, but for the fourteen families from Tetovo the difference is purely theoretical.

The same week we learned that 1.6 million euros' worth of granite tiles are going down on a car park for MPs, on land a company bought at public auction for 333,000 euros before the law stopped it from building there. Compensation is not mentioned in a single statement. And an 82-year-old man in Veles was detained for beating his wife for years - the system responded, but it responded last.

What ties these stories together is not idleness. There was plenty of work. The problem is that what got measured everywhere was the institution's own motion - how many sprayings, how many solutions drawn up, how many investigations opened - while the number that actually means something, the one counting the sick, the flooded and the families still without a verdict, moved the other way. The average wage fell to 48,381 denars, another figure nobody counts out loud. Measurement became a substitute for the result, and it is a comfortable substitute precisely because it can be verified - while the result cannot.